About 40% of all prescribed drugs are cationic at physiological pH and the multidrug Organic Cation Transporters, OCT1 and OCT2, are the initial steps in their hepatic and renal secretion, respectively. Their broad selectivity makes both of these processes principal sites for unwanted drug-drug interactions (DDIs). The impact these transporters have on DDIs, as well as the influence they have on the pharmacokinetics of cationic drugs, makes them a focus of efforts to develop models capable of predicting and, ideally, pre-empting unwanted interactions. Moreover, FDA guidance increasingly urge that New Molecular Entities (NMEs) be tested using defined protocols ('Decision Trees') designed to identify potentials DDIs with OCTs, as well as other multidrug transporters, and the results influence key decisions on the clinical and commercial future of these compounds. The value of these efforts (for both modeling and regulatory recommendations) is, however, compromised by a lack of understanding of two critical issues: (1) we cannot predict with sufficient quantitative accuracy the extent to which novel drugs or preclinical molecules will interact with either OCT1 or OCT2; and (2) we have no theoretical (or empirical) basis upon which to predict whether an inhibitor of OCT activity is itself a substrate for transport. The present proposal describes studies that directly address both these issues by confronting two critical flaws in previous efforts to model drug interactions with OCT1 and OCT2, i.e., the failure to acknowledge the influence of 'substrate' on the 'selectivity' profiles upon which current models of drug interactio with polyspecific OCTs have been based; and the absence of accurate information on the structures of these proteins. To fill these gaps in knowledge we will integrate ligand-based and protein- based approaches for development of structure/function relationships for human OCT1 and OCT2. The plan is organized around two major aims. Aim 1 will develop models of drug interaction with OCT1 and OCT2 using newly developed protocols capable of high throughput and high resolution assessment of the mechanistic basis of substrate-specific ligand interaction with these proteins. These protocols include a means to determine if inhibitors of OCT activity are themselves substrates for transport. We will also introduce a way to share the models we develop via web- and mobile device-based applications. Aim 2 will determine the crystal structure of OCT1 and OCT2, thereby clarifying the structural basis of ligand-interaction with OCTS, including the influence of polymorphisms on substrate- and inhibitor-specific activity profiles. In summary, the proposed program will develop robust computational models of ligand interaction with OCTs that, when integrated with knowledge of the crystal structures of OCT transporters, will increase opportunities to identify novel OCT substrates and inhibitors, thereby holding the promise of assisting in drug development as well as predicting and preempting adverse drug interactions.